An electrical structure that has both an input and an output is represented as a linear two-port network or circuit. The input of the two-port circuit is assumed to be fed or sourced with a two-wire pair, and likewise, the output of the two-port circuit is loaded or terminated with a two-wire pair. The two-wire input pair and the two-wire output pair sometimes have a ground conductor in common.
At high frequencies, voltages and currents are derived from measurements of magnitude and phase of a signal wave traveling in a given direction. Thus, voltages and currents, and related impedance and admittance parameters, are not directly measurable. Instead, frequency-domain measurements, such as measurements of scattering or S parameters, are made with a network analyze, which is a type of test equipment. Once the scattering parameters of the network are known, conversion to other parameters, such as impedance or admittance parameters, can be performed if desired.
S parameters are discussed in relation with FIG. 1, which shows a two-port network having port 1 and port 2. The xe2x80x9caxe2x80x9d parameter is a normalized incident wave (i.e., the wave that is input to a port) and the xe2x80x9cbxe2x80x9d parameter is a reflected wave (i.e., the wave is output from the port). The S parameters are dimensionless. Port 1 is associated with incident and reflected waves a1, and b1, while port 2 is associated with waves a2 and b2. The value |a1|2 represents the power incident on port 1, the value |b1|2 represents the power reflected from port 1, the value |a2|2 represents the power incident on port 2, and the value |b2|2 represents the power reflected from port 2.
Given the two-port network of FIG. 1, the relation between a1, a2, and b1, b2 is as follows:
b1=S11a1+S12a2, and
b2=S21a1+S22a2.
More generally, the incident wave matrix [a] is related to the reflected wave matrix [b] by the scattering matrix [S]:
[b]=[S] [a].
A communications link may have multiple components, such as a coaxial cable, a connector, a board trace, and so forth. Each component of the communications link has its impact on the overall performance of the link. The components when connected in series can be treated as cascaded networks, with each network representing a corresponding component. While certain components can be tested directly by a network analyzer, testing of other components require the use of test fixtures if the network analyzer has test ports that may not adequately test such other of components.
A test fixture is usually a circuit board that is configured for testing a device under test (DUT), which can be one of the components of a communications link. When a test fixture is introduced, its contribution to the overall measurement made by the test equipment needs to be separated to accurately determine performance of the device under test. A limitation of S parameters is that separation of test fixture effects from the measured S-parameter data is a difficult task.
An improved mechanism is provided for separating test fixture effects from measured scattering (or S) parameter data for multi-port networks. For example, a method of characterizing a device under test includes receiving measured frequency-domain parameters, such as scattering (or S) parameters, of a circuit including a test fixture and the device under test. The scattering parameters are transformed to a first set of further network parameters (such as transmission or T parameters). The effect of the test fixture is separated from the first set of T parameters to derive a second set of T parameters to represent performance of the device under test.
Other or alternative features will become apparent from the following description, from the drawings, and from the claims.